Zeema
New design
Catalogue / api-gateway

Unify

API gatewayunify-ai

Hosted LLM router that chose a model and provider per prompt from live quality, cost and speed benchmarks.

Provider
default
Kind
API gateway
Category
Edge
Official docs
https://github.com/unifyai/docs
Properties · 2 set by this component
propertytypevaluesrequiredkind defaultthis component
authenumnone | api-key | jwt | oauth2 | mtlsnononeinherited
concurrencyLimitboolean—nofalseinherited
idempotencyKeys · advancedboolean—nofalseinherited
instancesinteger—no—inherited
keyTtlHours · advancedinteger—no—inherited
managedboolean—nofalse▸ true
protocolenumhttp | grpc | websocket | fix | ouchnohttp▸ http
rateLimitingboolean—nofalseinherited
selfHosted · component propertyboolean—nofalseinherited
tlsboolean—notrueinherited
Rules that apply · 1

Every rule whose selector covers this component's kind or its category. Whether a rule fires depends on the design: its properties, its edges and its groups.

Synchronous call chain too deepwarning
sync-chain-too-deep

Every consecutive synchronous hop from this entry point spends part of the stated p95 latency budget, and the chain's failures multiply. More than three hops is a budget nobody is tracking.

  • — all of these hold:
  • — the design requirement `latency.p95Ms` has a value
  • — the longest chain of synchronous calls starting at the node is more than 3 hops long
Use it in a design
nodes:
  - { id: "unify-ai", component: "unify-ai" }
Other API gateway products · 40
API Gatewayalibaba-api-gatewayalibabacloudapi-gatewayAmazon API Gatewayaws-api-gatewayawsapi-gatewayAWS AppSync GraphQL APIaws-appsyncawsapi-gatewayAzure API Managementazure-api-managementazureapi-gatewayAI Gatewaycloudflare-ai-gatewaycloudflareapi-gatewayCloudflare API Shieldcloudflare-api-shieldcloudflareapi-gatewayMCP Server Portalcloudflare-mcp-portalcloudflareapi-gatewayCloudflare Privacy Gatewaycloudflare-privacy-gatewaycloudflareapi-gatewayGeneric API gatewayapi-gatewayapi-gatewayApache APISIXapisixapi-gatewayApollo GraphOSapollo-graphosapi-gatewayEasegresseasegressapi-gatewayEmissary-ingressemissaryapi-gatewayEnvoy Gatewayenvoy-gatewayapi-gatewayGloo Gatewayglooapi-gatewayHasurahasuraapi-gatewayHelicone AI Gatewayhelicone-gatewayapi-gatewayHigresshigressapi-gatewayKong Gatewaykongapi-gatewayKong AI Gatewaykong-ai-gatewayapi-gatewayKong Konnectkong-konnectapi-gatewayKrakenDkrakendapi-gatewayLiteLLM Proxylitellmapi-gatewayMartian Gatewaymartianapi-gatewayNot Diamondnot-diamondapi-gatewayPortkey AI Gatewayportkeyapi-gatewayRequestyrequestyapi-gatewaySpring Cloud Gatewayspring-cloud-gatewayapi-gatewayTyk Gatewaytykapi-gatewayVercel AI Gatewayvercel-ai-gatewayapi-gatewayWSO2 API Managerwso2-api-managerapi-gatewayZuulzuulapi-gatewayApigeeapigeegcpapi-gatewayAgent Gatewaygoogle-agent-gatewaygcpapi-gatewayAPI Gatewaygoogle-api-gatewaygcpapi-gatewayApigee hybridgoogle-apigee-hybridgcpapi-gatewayCloud Endpointsgoogle-cloud-endpointsgcpapi-gatewayFirebase Data Connectgoogle-firebase-data-connectgcpapi-gatewayIBM API Connectibm-api-connectibmapi-gatewayOCI API Gatewayoci-api-gatewayociapi-gateway
Source
# yaml-language-server: $schema=../../schema/component.schema.json
# Unify router: hosted LLM router that picked a model and provider per prompt from live benchmarks.
# Retired September 2026: Unify pivoted to AI assistants and then agent post-training, docs.unify.ai went
# offline and its source was archived on GitHub ("unifyai/docs", "the documentation site ... 2024-2026").
# No surviving page documents router defaults, so only the hosted shape is recorded.
schemaVersion: "1"
type: "component"
id: "unify-ai"
kind: "api-gateway"
provider: "default"
docs: "https://github.com/unifyai/docs"
defaults:
  managed: true
  protocol: "http"
properties:
  selfHosted:
    type: "boolean"
    default: false

The same page as markdown, for an agent: /catalogue/components/unify-ai.md